Background: Approximately 30% of TNBCs are characterized by microarray as claudin-low, mesenchymal or mesenchymal stem cell-like and, unlike basal TNBCs, these tumors frequently harbor aberrations in the PI3K/AKT/mTOR axis, raising the possibility of targeting this axis to enhance chemotherapy response. Assays to clinically identify mesenchymal TNBCs are under development, but published results confirm that up to 30% are metaplastic breast cancers (MpBCs), a chemo-refractory group of tumors that contain a mixture of epithelial and mesenchymal components, making them identifiable by microscopy. As such, MpBCs serve as surrogates of response for potential regimens to treat mesenchymal TNBC. Methods: Patients (pts) with advanced TNBC (N=64) were treated with liposomal doxorubicin (D), bevacizumab (A) and the mTOR inhibitors temsirolimus (T) or everolimus (E). D and A were administered IV on day 1 with T (IV on days 1, 8 and 15) or E (continuous daily oral administration) using 21 day cycles. Response was assessed every 6 weeks using RECIST. When available, archived tissue was evaluated for aberrations in the PI3K pathway using standard assays. Results: Fifty-two MpBC pts were treated with DAT (N=39) or DAE (N=13). Median age was 58 (range 37-79); median # of prior regimens for metastatic disease was 1 (range 0-5). The objective response rate (ORR) was 21% [complete response (CR)=4 (8%); partial response (PR)=7 (13%)] and 10 (19%) pts had stable disease (SD)≥6 months for a clinical benefit rate (CBR) of 40%. Tissue was available for testing in 43 pts and 32 (74%) had a PI3K pathway activating aberration (Table 1). Response According to PI3K Pathway AberrationPI3K Pathway AberrationN (%)CRPRSD≥6monthsCBRORRAny PI3K Pathway Aberration*32 (74)46444%31%PIK3CA Mutation19 (59)23447%26%p.H1047R12 (38)21350%25%p.E545K6 (19)02150%33%p.G1007R1 (3)010100%100%p.E545A1 (3)0000%0%p.H1047Y1 (3)0000%0%p.K111E1 (3)0000%0%p.E542K1 (3)0000%0%PIK3CA Amplification1 (3)010100%100%PTEN Mutation5 (16)0000%0%PTEN Loss5 (16)02040%40%AKT1 p.E17K Mutation2 (6)0000%0%AKT2 Amplification1 (3)100100%100%PIK3R1 Mutation2 (6)01050%50%NF2 Mutation1 (3)100100%100%No PI3K Pathway Aberration11 (26)00545%0%*Some tumors had >1 aberration detected PI3K pathway activation was associated with a significant improvement in ORR (31 vs 0%; P=0.043) but not CBR (44 vs 45%; P=1.000) or progression-free survival (median 5.1 vs 2.9 months; P=0.352). A pt with 5 year+ durable CR (on maintenance everolimus) had a mutation in NF2. To emphasize the importance of pt selection, it is notable that 12 pts with non-metaplastic TNBC were also treated with DAT, and only 1 pt had a response (CR/PR=1; SD≥6 months=0), for a CBR that was significantly worse than pts with MpBC (8 vs 40%; P=0.045). Conclusions: Using MpBC as a surrogate of response, DAT/DAE has significantly better activity in mesenchymal compared to non-selected TNBC. Response is enhanced in pts with PI3K pathway activation. DAT/DAE should be tested in non-metaplastic, mesenchymal TNBC once a diagnostic assay is available. Citation Format: Basho RK, Gilcrease M, Murthy RK, Helgason T, Booser DJ, Karp DD, Meric-Bernstam F, Wheler JJ, Valero V, Albarracin C, Litton J, Chavez-MacGregor M, Ibrahim NK, Murray JL, Koenig KB, Hong D, Subbiah V, Kurzrock R, Janku F, Moulder S. Targeting the PI3K/AKT/mTOR pathway for the treatment of mesenchymal triple-negative breast cancer (TNBC): Evidence of efficacy and proof of concept from a phase I trial with dose expansion of mTOR inhibition in combination with liposomal doxorubicin and bevacizumab. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-14-02.

Background: HER2-targeted therapies have improved survival for advanced HER2-positive breast cancers (BC), but none have been approved for tumors with low levels of HER2 expression (ie, HER2 IHC 1+ or 2+/ISH-negative). Trastuzumab deruxtecan (DS-8201a) is a novel HER2-targeted antibody-drug conjugate with a humanized HER2 antibody attached to a potent topoisomerase I inhibitor payload by a cleavable peptide-based linker, which is designed to have broad antitumor activity in HER2-expressing tumors. It has a drug-to-antibody ratio of 7 to 8, a novel linker that is stable in plasma and that is selectively cleaved by lysosomal cathepsins which are upregulated in cancer cells, and its payload has a short systemic half-life. In 2015, a phase 1 study (NCT02564900) was initiated to evaluate the safety and efficacy of DS-8201a in subjects with advanced HER2-expressing or HER2-mutated solid tumors, including HER2-low expressing BC. In this study, the overall confirmed response rate (ORR) in the evaluable subjects was 49.3% (103/209) (April 2018 data cutoff; Iwata H, et al. ASCO 2018). Expanded results from the HER2-low expressing BC subjects are presented here. Methods: This ongoing phase 1 trial included 2 parts. The dose escalation part served to determine the dose-limiting toxicities, the maximum tolerated dose, and to select the recommended dose for expansion (RDE). The dose expansion part further evaluated the safety, tolerability, and efficacy of the DS-8201a at the RDE (5.4 and 6.4 mg/kg; q3wks) in various advanced HER2-expressing or HER2-mutated solid tumors, including heavily pretreated HER2-low BC (IHC 1+ or 2+, ISH-negative). Enrollment of HER2-low subjects is ongoing. Results: At the cutoff date of 18 April 2018, data from 34 HER2-low BC subjects were collected. The median age was 55.8 (range; 33, 75) years, the median number of prior endocrine therapies was 2, and the median number of prior chemotherapies was 3. In this HER2-low BC population, most patients had hormone receptor (HR)-positive disease (85.3%; 29/34); of which 17.2% (5/29) received prior treatment with a CDK4/6 inhibitor. The confirmed ORR was 50.0% (17/34), the disease control rate was 85.3% (29/34), the median time to response was 2.8 (range; 1.2, 13.8) months, the median duration of response (DOR) was 11.0 months, and the median progression-free survival (PFS) was 12.9 months. In the subgroup with HR-positive disease, the ORR was 55.2% (16/29), the median DOR was 11.0 months, and the median PFS was 13.6 months. After exclusion of 8 HER2-low subjects who received prior HER2-targeted therapy, the ORR was 46.2% (12/26). In the overall study, among the 145 BC subjects who received ≥1 dose of DS-8201a (5.4 or 6.4 mg/kg), the most frequent grade ≥3 adverse events included anemia (14.5%), and decreased counts of neutrophils (13.8%) and white blood cells (10.3%). There were 4 fatal cases of interstitial lung disease/pneumonitis in BC subjects, including 2 fatal cases in HER2-low BC subjects. Conclusions: In this study, DS-8201a showed substantial antitumor activity and acceptable safety in heavily pretreated HER2-low BC. Citation Format: Modi S, Tsurutani J, Tamura K, Park H, Sagara Y, Murthy R, Iwata H, Krop IE, Doi T, Redfern C, Moreno-Aspitia A, Redman R, Lee C, Sugihara M, Fujisaki Y, Takahashi S. Trastuzumab deruxtecan (DS-8201a) in subjects with HER2-low expressing breast cancer: Updated results of a large phase 1 study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-17-02.